ham
David Wilson
davidwwilson at comcast.net
Tue May 10 13:24:07 CEST 2005
You will notice that my posts sometimes often include "ham" among the
recipients. This will result in a bounce if you "reply all" to my posts,
and several people have commented on this.
By way of explanation:
I am currently using MicroSuck Outlook Express, the free but rather hobbled
version of Outlook that came on my PC.
About a year ago, my spam ratio finally exceeded my patience ratio. Since
Outlook Express does not provide an effective spam filter, I decided to try
Spambayes, of which I had heard good things (which were true).
Most mail programs provide API hooks allowing Spambayes filtering to be
integrated into the program. Not so Outlook Express. When using Spambayes
with Outlook Express, Spambayes must be inserted upstream of Outlook
Express, where it marks incoming mail so that Outlook Express can classify
it using its (lame) native mail filters. From a small number of unsavory
choices, Spambayes chose to include "ham", "spam" or "unsure" in the "To:"
line, so that Outlook Express could identify and route spam by detecting
"spam" in the "To:" line. I am guessing that this Spambayes feature was
included solely to deal with Outlook Express.
Unfortunately, because good incoming mail has "ham" in the "To:" line, "ham"
is replicated to the "To:" line of reply-alls to that mail. If I fail to
remove "ham" from the recipients, it makes its way into the mailing list and
subsequent "reply-alls". Since "ham" is not a valid recipient, bounces
occur.
I don't know if Seqfan and Math-Fun could filter "ham" out of the "To:" line
of incoming mail, this would solve the mailing list problem. A more general
solution would be for Spambayes to allow me to put "spam" in the "To:" line
of spam, but leave the "To:" line of ham alone; this is consistent with my
own personal mail filtering.
- David W. Wilson
"Truth is just truth -- You can't have opinions about the truth."
- Peter Schickele, from P.D.Q. Bach's oratorio "The Seasonings"
More information about the SeqFan
mailing list